Orthogonal control of protein function allows one protein from a family of related proteins to be inactivated so that its unique biological function can be assessed. We propose development of a new kind of orthogonal control, called allosteric switches, in the caspase family. Caspases are cysteine proteases that execute cell death by cleaving a discrete selection of target proteins, which ultimately leads to apoptosis. Caspases have a large central cavity that can be allosterically inhibited by small molecules. Caspases are believed to be good targets for treatment of cancer (caspase activators) or stroke, heart attack and Alzheimer's Disease (caspase inhibitors). Unfortunately the similarities in currently available small inhibitors that bind at the caspase active sites have not allowed unambiguous identification of the precise roles of the twelve different caspases. We will use a combination of directed evolution and computational protein design to introduce an allosteric switch in caspases-3, -6, and -7. When the small molecule effector we have selected or designed against binds to the evolved allosteric site, caspase activity will be inhibited. This switch will allow us to turn off one type of caspase at a time to determine what proteins are specifically cleaved by caspase-3 or -6 or -7. The cleaved proteins themselves are inherently interesting, because their cleavage is proapoptotic, meaning they lead to cell death. The ability to induce apoptosis is the hallmark of a useful cancer drug. Our studies may suggest which caspase should be targeted for treating which disease. Successful implementation of allosteric switches in these caspases will pave the way for allosteric switches in the other nine caspases and ultimately into other families of proteins where biological information is lacking. NON-TECHNICAL SUMMARY: The central problem in disease treatment is knowing which protein should be targeted to treat which disease. The allosteric switch technology we will develop in caspases allows just one type of protein to be turned off by a drug, to determine which disease could be effectively treated by targeting that particular protein. We focus on caspase proteins, which can cause cancer cells to die.